Value Of Multimodality Magnetic Resonance Imaging In Grading Glioma And Genotyping IDH

Background and Objective:Gliomas,defined as tumors arising from brain glial cells,are the most common primary intracranial tumors.The incidence of glioma is 5-6 cases/100,000,accounting for about 24%of adult primary brain tumors;the World Health Organization（WHO）classification of central nervous system tumors divides glioma into grades I～IV,grades I and II are low grade glioma（LGG）,and grades III and IV are high grade glioma（HGG）.The treatment strategies adopted for different grades of gliomas in clinical practice are significantly different,so the accurate assessment of preoperative glioma grade is of great clinical significance for the selection of treatment options.Genotyping was first added to gliomas by the 2016WHO classification of central nervous system tumors,highlighting the importance of molecular diagnosis in glioma treatment and prognosis.Isocitrate dehydrogenase（IDH）is the most common gene mutation type in glioma.The survival prognosis of patients with IDH mutant glioma and IDH wild-type glioma varies greatly.Therefore,preoperative detection of IDH genotype in glioma patients is essential for the development of treatment options and prognosis evaluation of patients.Surgery is the gold standard for obtaining glioma pathology and gene type,but it is not only invasive but also has some limitations for patients who cannot be treated surgically.At present,although there are a variety of magnetic resonance imaging（MRI）techniques that can be used for the pathological grading and gene type diagnosis of glioma,conventional MRI techniques have many shortcomings,so exploring new noninvasive MRI methods for the pathological grading and gene type of glioma has been the focus of research.In this study,multimodality MRI was used to evaluate the pathological grade and IDH genotyping of gliomas in order to provide an important reference for the development of treatment strategies and prognostic evaluation of glioma patients.Materials and methods:Thirty-six glioma patients,including 22 males and 14 females,with a mean age of 52.8±10.7 years（range:23 to 76 years）,who visited Henan Provincial People’s Hospital from July 2020 to November 2021 were prospectively collected.All subjects underwent magnetic resonance elastography（MRE）,diffusion kurtosis imaging（DKI）,diffusion weighted imaging（DWI）,and arterial spin labeling（ASL）sequences on a GE 3.0T magnetic resonance imager.MRE,DWI,DKI,and ASL sequences were post-processed on a GE functool ADW4.6 workstation,and the post-processing results of each sequence were independently recorded by two physicians who were completely unaware of the patient’s pathological grade and IDH genotyping,including the maximum shear stiffness(Stiffness_max),minimum shear stiffness(Stiffness_min)and mean shear stiffness(Stiffness_mean)of the MRE sequence,apparent diffusion coefficient（ADC）,slow diffusion coefficient（ADCslow）,rapid diffusion coefficient（ADCfast）,perfusion fraction（f）,distribution diffusion coefficient（DDC）and heterogeneity index（α）of the DWI sequence,mean kurtosis（MK）,mean diffusion rate（MD）as well as fractional anisotropy（FA）of the DKI,and cerebral blood flow（CBF）of the ASL sequence.First,all measurement data were tested for normality using the Shapiro-Wilk test test.Measurement data that conformed to the normal distribution were expressed as（?）±SD.Independent sample t-test was used for the difference between high and low grade glioma and IDH wild-type and mutant glioma.If the normal distribution was not met,Mann-Whitney U test was used for the difference between high and low grade glioma and IDH wild-type and mutant glioma.Pearson chi-square test was used for the difference of categorical variables such as gender and age between high and low grade glioma and IDH wild-type and mutant glioma.Receiver operating characteristic（ROC）curve analysis was used to assess the diagnostic efficacy of each image parameter for pathological grade and IDH genotyping,and the area under the curve（AUC）and corresponding 95%confidence interval（CI）were calculated.Intraclass correlation coefficient（ICC）was used to assess the agreement between the two radiologists,and ICC>0.75 was considered good agreement between the two radiologists.Results:1.General resultsOf the 36 glioma patients,21 had high-grade glioma（10 grade III and 11 grade IV）and 15 had low-grade glioma（grade II）;of the 36 glioma patients,24 had IDH mutant glioma,including 13 high-grade and11 low-grade;and 12 had IDH wild-type glioma,including 8 high-grade and 4 low-grade.There were no statistically significant differences in the age and gender of the patients between high and low grade gliomas and between IDH mutant and wild-type gliomas（P>0.05）.2.Evaluation results of MRE,DWI,DKI and ASL for pathological grading of gliomaCompared with low-grade gliomas,the Stiffness_meanof MRE in high-grade gliomas was significantly lower（P<0.05）,with an AUC of 0.968,and its sensitivity and specificity were 93.3%and 90.5%respectively when Stiffness_meantakes 1861.0 as cutoff value,while there was no statistical significance in Stiffness_minand Stiffness_maxbetween the two groups（P>0.05）;Compared with low-grade gliomas,the ADC,ADCslow,f,and DDC of high-grade gliomas were significantly lower（P<0.05）,the AUC of ADCslow was higher at 0.895,and its sensitivity and specificity were 80.0%and 85.7%respectively when ADCslow takes 0.594 as cutoff value,while there was no statistical significance in ADCfast andαbetween the two groups（P>0.05）;Compared with low-grade gliomas,the MK was significantly higher and MD was significantly lower in high-grade gliomas（P<0.05）,the AUC of MD was higher at 0.913,and its sensitivity and specificity were 86.7%and 76.2%respectively when MD takes 0.861 as cutoff value,while there was no statistical significance in FA between the two groups（P>0.05）;Compared with low-grade gliomas,high-grade gliomas had significantly higher CBF values（P<0.05）,with an AUC of 0.711,and its sensitivity and specificity were 90.5%and 60.0%respectively when CBF takes 38.285 as cutoff value.3.Evaluation results of IDH genotyping of glioma by MRE,DWI,DKI and ASLCompared with IDH mutant glioma,Stiffness_meanand Stiffness_maxof MRE were significantly lower in IDH wild-type glioma（P<0.05）,the AUC of Stiffness_meanwas higher at 0.929,and its sensitivity and specificity were 91.7%and 75.0%respectively when Stiffness_meantakes 1555.5 as cutoff value,while there was no statistical significance in Stiffness_minbetween the two groups（P>0.05）;Compared with IDH mutant glioma,the ADCslow,f were significantly lower and the ADCfast was significantly higher（P<0.05）,the AUC of f was higher at 0.861,and its sensitivity and specificity were 87.5%and 58.3%respectively when f takes 0.57 as cutoff value,whlie there was no statistical significance in ADC,αand DDC between the two groups（P>0.05）;Compared with IDH mutant glioma,the MD was significantly lower in IDH wild-type glioma（P<0.05）,with an AUC of 0.875,and its sensitivity and specificity were79.2%and 83.3%respectively when the MD takes 1.13 as cutoff value,while there was no statistical significance in MK and FA between the two groups（P>0.05）;Compared with IDH mutant gliomas,IDH wild-type gliomas had significantly higher CBF values（P<0.05）,with an AUC of 0.809,and its sensitivity and specificity were 66.7%and 79.2%respectively when CBF takes 63.221 as cutoff value.Conclusions:1.MRE,DWI,DKI,and ASL can be used for preoperative glioma pathological grading,with the highest diagnostic efficacy of Stiffness_meanfor MRE,which has an AUC of 0.968,and its sensitivity and specificity were 93.3%and 90.5%respectively when Stiffness_meantakes 1861.0 as cutoff value.2.MRE,DWI,DKI,and ASL can be used to assess the IDH gene type of preoperative gliomas,with the highest diagnostic efficacy of Stiffness_meanfor MRE,which has an AUC of 0.929,and its sensitivity and specificity were 91.7%and 75.0%respectively when Stiffness_meantakes 1555.5 as cutoff value.3.Multimodality MRI can provide rich information from different perspectives for the evaluation of pathological grade of glioma and IDH gene type,and provide important value for the development of treatment strategies and prognostic evaluation of glioma patients.